AI Engineer
Core
Build retrieval, matching, and LLM systems for AI research products that connect scientific papers to video content and provide cited answers.
Role type
Senior IC machine-learning engineer (RAG, search, LLM)
Builds
Hybrid retrieval systems, paper-to-video matching pipelines, RAG/agentic answer synthesis, protocol extraction tools, and production LLM APIs.
Domain
Scientific research, education, video content, large-scale document retrieval
Deliverable
production ML models
Required skills
Python, software engineering (testing, CI/CD), RAG systems (chunking, embeddings, reranking), vector/search engines (Elasticsearch, Milvus, pgvector, Qdrant, Pinecone), LLM APIs (Anthropic, OpenAI, Hugging Face, vLLM), offline evaluation, agent frameworks, cloud infrastructure (AWS), Docker, API design
Preferred skills
Scholarly data experience (PubMed, OpenAlex, PMC), multimodal work (video transcripts, timestamp alignment), model fine-tuning, large-scale indexing
Technologies
Python, Elasticsearch, OpenSearch, Milvus, pgvector, Qdrant, Pinecone, Anthropic, OpenAI, Hugging Face, vLLM, AWS, Docker
Responsibilities
Build hybrid retrieval with reranking over tens of millions of papers; improve paper-to-video matching; build RAG and agentic pipelines for cited synthesis; extract method steps from papers; build evaluation sets and track metrics; ship and run models in production; build APIs and MCP servers; measure search and video funnels
Seniority
Senior, hands-on IC
